A Vehicle Speed Estimation Using Faster RCNN

Abstract

The transport network, especially the road system, has faced huge pressures due to increased population growth and urbanization. One of the negative side effects of traffic growth is road accidents. Intelligent Transportation system (ITS) is like the Human vision system (HVS).  Deep learning-based classification and detection algorithms have emerged as powerful tools for vehicle detection in intelligent transportation systems. The limitation of the number of high-quality labeled training samples makes the single vehicle detection in intelligent transportation systems. This paper presents the detection and tracking of vehicles on the recorded dataset by utilizing the Faster RCNN architecture and Matlab 2023 programming language. After training the object detection model using Faster RCNN the detection accuracy 100% and then applying Kalman Filter to track the vehicles detected by the Faster RCNN. The system has proven its effectiveness in detecting and distinguishing vehicles.